EUVIMEDEUVIMED entdecken
Crossref

Machine learning-driven clinical decision support for low bone mineral density: A web-based prediction model with explainable AI integration

Xing Yang, Jianyuan Liu, Xiaozhi Huang, Hao Liang, Ping Cui, Shiran He, Heng Zhang, Wenping Liao, Guangkun Zhang, Qianqian Huang, Huan Ning, Tingyan Luo und 3 weitere

Bone · 2025-11 · 200 · 117592

DOI: 10.1016/j.bone.2025.117592 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bone, 2025-11. Verfasst von Xing Yang und 14 weiteren Autor:innen. Beteiligte Einrichtung: Health Management Research Institute, People's Hospital of Guangxi Zhuang Autonomous Region and Guangxi Academy of Medical Sciences, Nanning, China. Hauptthemen laut MEDLINE: Machine Learning, Bone Density, Decision Support Systems, Clinical, Internet. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 46 Literaturangaben. Gefördert von: Natural Science Foundation of Guangxi Zhuang Autonomous Region, Science and Technology Major Project of Guangxi.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND: Low bone mineral density (LBMD), which includes osteopenia and osteoporosis, is associated with substantial health care costs. However, current diagnostic methods for LBMD are limited in terms of accuracy and accessibility. This study aims to develop an interpretable machine learning model for LBMD risk assessment and implement it as a web-based clinical decision support tool. METHODS: Data from subjects who underwent dual-energy X-ray absorptiometry (DXA) at the People's Hospital of Guangxi Zhuang Autonomous Region were collected and randomly divided into a training set (70 %) and an internal validation set (30 %). An external validation set was sourced from the National Health and Nutrition Examination Survey (NHANES) database. Least absolute shrinkage and selection operator (LASSO) regression and multiple logistic regression were used for feature selection. Ten common machine learning models were conducted based on the selected features. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), Matthews correlation coefficient (MCC), Brier score, and decision curve analysis (DCA). The decision mechanisms of the best-performing model were explained using SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME). The optimal model was deployed as a web application using Streamlit. RESULTS: A total of 16,274 participants were included in this study. Age, body mass index (BMI), alkaline phosphatase, and total cholesterol were identified as key predictors of LBMD. The logistic regression (LR) model demonstrated superior prediction performance (internal validation set [AUC = 0.902, MCC = 0.684, Brier score = 0.123], external validation set [0.812, 0.358, 0.265]). DCA confirmed its clinical utility. Both SHAP and LIME showed consistent results in identifying predictive factors. The LR model was deployed as a web application to predict LBMD. CONCLUSION: Our interpretable machine learning model and web-based implementation provide a free and reliable tool for predicting LBMD, which represents a significant advancement in making LBMD screening more accessible and cost-effective.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningBone DensityDecision Support Systems, ClinicalInternetHumansFemaleMaleMiddle AgedAbsorptiometry, PhotonAgedAdultROC CurveOsteoporosis

Stichwörter der Autor:innen

LIMELogistic regressionLow bone mineral densityMachine learningPrediction modelRisk factorsSHAP

Autor:innen und Institutionen

  • Health Management Research Institute, People's Hospital of Guangxi Zhuang Autonomous Region and Guangxi Academy of Medical Sciences, Nanning, China.
  • School of Public Health, Guangxi Medical University, Nanning, China; Guangxi Key Laboratory of AIDS Prevention and Treatment, Guangxi Medical University, Nanning, China.
  • Health Management Center, People's Hospital of Guangxi Zhuang Autonomous Region and Guangxi Academy of Medical Sciences, Nanning, China.
  • Guangxi Key Laboratory of AIDS Prevention and Treatment, Guangxi Medical University, Nanning, China; Life Sciences Institute, Guangxi Medical University, Nanning, China.
  • Health Management Center, People's Hospital of Guangxi Zhuang Autonomous Region and Guangxi Academy of Medical Sciences, Nanning, China. Electronic address: 26895185@qq.com.
  • Health Management Center, People's Hospital of Guangxi Zhuang Autonomous Region and Guangxi Academy of Medical Sciences, Nanning, China. Electronic address: liwei2186gxqyy@163.com.

Förderung

  • Natural Science Foundation of Guangxi Zhuang Autonomous Region
  • Science and Technology Major Project of Guangxi

Zitieren

APA

Yang, X., Liu, J., Huang, X., Liang, H., Cui, P., He, S., Zhang, H., Liao, W., Zhang, G., Huang, Q., Ning, H., Luo, T., Luo, Y., Li, W., & Huang, J. (2025). Machine learning-driven clinical decision support for low bone mineral density: A web-based prediction model with explainable AI integration. Bone, 200, 117592. https://doi.org/10.1016/j.bone.2025.117592

Vancouver

Yang X, Liu J, Huang X, Liang H, Cui P, He S, et al. Machine learning-driven clinical decision support for low bone mineral density: A web-based prediction model with explainable AI integration. Bone. 2025;200:117592. doi:10.1016/j.bone.2025.117592

BibTeX
@article{Yang2025,
  author = {Yang, Xing and Liu, Jianyuan and Huang, Xiaozhi and Liang, Hao and Cui, Ping and He, Shiran and Zhang, Heng and Liao, Wenping and Zhang, Guangkun and Huang, Qianqian and Ning, Huan and Luo, Tingyan and Luo, Yinghua and Li, Wei and Huang, Jiegang},
  title = {Machine learning-driven clinical decision support for low bone mineral density: A web-based prediction model with explainable AI integration},
  journal = {Bone},
  year = {2025},
  volume = {200},
  pages = {117592},
  doi = {10.1016/j.bone.2025.117592}
}
Kostenlos starten →